CN110360725B - Air conditioner, cloud server, and method for driving and controlling air conditioner - Google Patents

Air conditioner, cloud server, and method for driving and controlling air conditioner Download PDF

Info

Publication number
CN110360725B
CN110360725B CN201910284184.1A CN201910284184A CN110360725B CN 110360725 B CN110360725 B CN 110360725B CN 201910284184 A CN201910284184 A CN 201910284184A CN 110360725 B CN110360725 B CN 110360725B
Authority
CN
China
Prior art keywords
operation mode
air conditioner
temperature
section
unit
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910284184.1A
Other languages
Chinese (zh)
Other versions
CN110360725A (en
Inventor
朴润植
权英卓
韩东雨
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
LG Electronics Inc
Original Assignee
LG Electronics Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by LG Electronics Inc filed Critical LG Electronics Inc
Publication of CN110360725A publication Critical patent/CN110360725A/en
Application granted granted Critical
Publication of CN110360725B publication Critical patent/CN110360725B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F1/00—Room units for air-conditioning, e.g. separate or self-contained units or units receiving primary air from a central station
    • F24F1/06—Separate outdoor units, e.g. outdoor unit to be linked to a separate room comprising a compressor and a heat exchanger
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/0008—Control or safety arrangements for air-humidification
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
    • F24F11/46—Improving electric energy efficiency or saving
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/50—Control or safety arrangements characterised by user interfaces or communication
    • F24F11/56—Remote control
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/50—Control or safety arrangements characterised by user interfaces or communication
    • F24F11/56—Remote control
    • F24F11/58—Remote control using Internet communication
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/50—Control or safety arrangements characterised by user interfaces or communication
    • F24F11/61—Control or safety arrangements characterised by user interfaces or communication using timers
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F24F11/64—Electronic processing using pre-stored data
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
    • F24F11/63—Electronic processing
    • F24F11/65—Electronic processing for selecting an operating mode
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/70—Control systems characterised by their outputs; Constructional details thereof
    • F24F11/72—Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure
    • F24F11/74—Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure for controlling air flow rate or air velocity
    • F24F11/77—Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure for controlling air flow rate or air velocity by controlling the speed of ventilators
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/70—Control systems characterised by their outputs; Constructional details thereof
    • F24F11/72—Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure
    • F24F11/79—Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure for controlling the direction of the supplied air
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/70—Control systems characterised by their outputs; Constructional details thereof
    • F24F11/80—Control systems characterised by their outputs; Constructional details thereof for controlling the temperature of the supplied air
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F11/00—Control or safety arrangements
    • F24F11/70—Control systems characterised by their outputs; Constructional details thereof
    • F24F11/80—Control systems characterised by their outputs; Constructional details thereof for controlling the temperature of the supplied air
    • F24F11/86—Control systems characterised by their outputs; Constructional details thereof for controlling the temperature of the supplied air by controlling compressors within refrigeration or heat pump circuits
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
    • G05B13/027—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion using neural networks only
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F2110/00—Control inputs relating to air properties
    • F24F2110/10—Temperature
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F2110/00—Control inputs relating to air properties
    • F24F2110/20—Humidity
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F2120/00—Control inputs relating to users or occupants
    • F24F2120/10—Occupancy
    • F24F2120/12—Position of occupants
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
    • F24F2120/00—Control inputs relating to users or occupants
    • F24F2120/20—Feedback from users
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
    • Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
    • Y02B30/70—Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating

Abstract

The present invention relates to an air conditioner, a cloud server, and a method for driving and controlling the air conditioner based on parameter learning using artificial intelligence, the air conditioner according to an embodiment of the present invention includes: a parameter generation unit for calculating one or more parameters in the fast operation mode; and an operation mode control unit controlling the air blowing unit or the outdoor unit based on operation mode information in which a comfortable operation mode is set after an interval in which the air blowing unit or the outdoor unit is operated in a fast operation mode, the fast operation mode being operated only within a preset time range, and a central control unit instructing the operation mode control unit to operate in the comfortable operation mode after the fast operation mode.

Description

Air conditioner, cloud server, and method for driving and controlling air conditioner
Technical Field
The present invention relates to an air conditioner, a cloud server, and a method for driving and controlling the air conditioner based on parameter learning.
Background
An air conditioner is provided to create a comfortable indoor environment, and provides people with a more comfortable indoor environment by discharging cool air into a room to adjust the indoor temperature and purify the indoor air.
Generally, an air conditioner includes an indoor unit installed indoors and an outdoor unit configured by a compressor, a heat exchanger, and the like to supply a refrigerant to the indoor unit.
On the other hand, the indoor unit and the outdoor unit of the air conditioner can be controlled separately. In addition, the outdoor unit of the air conditioner may be connected to at least one indoor unit, and may be operated in a cooling or heating mode by supplying a refrigerant to the indoor unit according to a requested operation state.
Recently, for the convenience of users, a technique of controlling an air conditioner according to a preferred temperature suitable for the user when controlling the air conditioner has been proposed. The applicant of the present invention will observe in more detail in fig. 1 the mechanism of operation of an air conditioner in a fast mode and a comfort mode.
Fig. 1 is a diagram illustrating an operation process of an air conditioner operating in a fast mode and a comfort mode. When the air conditioner is turned on (S21), the air conditioner is operated in a fast mode (or a fast zone or a fast operation zone, which means a zone in which the air conditioner is operated in the fast mode) according to a predetermined set temperature Ts (S22). At this time, the set temperature Ts may be selected by a user, or may be automatically set according to a prescribed condition. The fast mode is a configuration for rapidly cooling (or heating) a space. Thereafter, in the fast mode, the air conditioner is operated at the maximum cooling capacity, thereby rapidly cooling (or heating) the indoor space. In this process, when the set temperature and the indoor temperature are compared (S23), if the set temperature is reached, the indoor humidity is checked (S24), and when the indoor humidity is equal to or more than a predetermined value, the dehumidification mode is operated (S26). On the other hand, when the indoor humidity is equal to or less than the predetermined value, the operation is performed in the comfort mode (S25).
Here, the comfort mode (or comfort zone or comfort operation zone, which refers to a zone in which the air conditioner operates in the comfort mode) refers to an operation based on a new set temperature (Tsa) higher than the set temperature Ts. Including operation of power saving functions for sensing the environment and adjusting the cooling load (cooling operation output) to maintain comfortable cooling (or heating). This is to initially operate in a fast mode in a state where the air conditioner is turned on, and then to prompt switching to the comfort mode when the temperature of the space reaches a predetermined temperature.
However, such a comfort mode and a quick mode have a limitation in that a change in the ambient environmental conditions cannot be dynamically applied in a predetermined state. There are various variables that affect the operation of the fast mode and the comfort mode according to the change of temperature, the number of indoor persons, humidity, etc., but there is a limitation in reflecting this.
Therefore, although the indoor units of the air conditioner are fixedly arranged to operate in a specific space, it is necessary to study a scheme of aggregating information derived in the process of operating a plurality of indoor units to improve the performance of the air conditioner. Therefore, in the present specification, a control method for enabling each indoor unit to operate in an optimal manner in a plurality of operation mode sections using information derived during operation of a plurality of indoor units, and an air conditioner to which the method is applied will be described.
Disclosure of Invention
In order to solve the above problems, the present specification provides an apparatus and method based on learning that learns parameters calculated in an operation section of an air conditioner so as to be able to efficiently operate in a section of the air conditioner divided into two or more operation modes.
In the present specification, there are provided an apparatus and a method based on learning, which calculate an optimum operation mode of an air conditioner by using parameters calculated by indoor units of a plurality of air conditioners as learning factors.
In the present specification, there is provided a learning-based apparatus and method for controlling the operation of a subsequent stage based on a parameter generated during a rapid temperature change of an indoor unit during operation.
Objects of the present invention are not limited to the above objects, and other objects and advantages of the present invention, which are not mentioned, can be understood by the following description, and the present invention can be more clearly understood by embodiments of the present invention. Further, it is apparent that the objects and advantages of the present invention can be achieved by the means set forth in the claims and combinations thereof.
An air conditioner according to an embodiment of the present invention includes: a parameter generation unit for calculating one or more parameters in the fast operation mode; and an operation mode control unit controlling the air blowing unit or the outdoor unit based on operation mode information in which a comfortable operation mode is set after an interval in which the air blowing unit or the outdoor unit is operated in a fast operation mode, the fast operation mode being operated only within a preset time range, and a central control unit instructing the operation mode control unit to operate in the comfortable operation mode after the fast operation mode.
In addition, the operation mode information of the air conditioner according to an embodiment of the present invention is a result factor that is received from the cloud server after the communication unit transmits the parameter to the cloud server or calculated in a built-in learning unit, the operation mode control unit sets the wind direction and the wind volume of the blowing unit using the operation mode information,
an air conditioner driven based on learning according to an embodiment of the present invention includes: a step in which the parameter generation means calculates one or more parameters in the fast operation mode; a step in which the operation mode control unit controls the air supply unit or the outdoor unit based on operation mode information for setting a comfortable operation mode after the section operated in the fast operation mode; and a step in which the rapid operation mode is operated only within a preset time range, and the central control unit instructs the operation mode control unit to operate in the comfort operation mode after the rapid operation mode.
The cloud server of an embodiment of the present invention includes: a communication unit which receives, from the plurality of air conditioners, one or more parameters calculated in the fast operation mode in correspondence with a set temperature set for each of the air conditioners, and transmits, in correspondence with the one or more parameters, operation mode information for setting the comfortable operation mode to each of the plurality of air conditioners; and a learning unit that receives a parameter of a first air conditioner of the plurality of air conditioners as a learning factor and outputs operation mode information for setting a comfortable operation mode of the first air conditioner after a section in which the first air conditioner is operated in the fast operation mode.
When the embodiment of the present invention is applied, the air conditioner may calculate an operation mode corresponding thereto using the parameter calculated during operation as a learning factor.
When the embodiment of the present invention is applied, the cloud server may calculate an operation mode suitable for each air conditioner after learning based on parameters provided by the plurality of air conditioners calculated during operation.
When an embodiment of the present invention is applied, the target attainment temperature can be maintained within a predetermined range based on a smaller power consumption amount per hour after the air conditioner is operated to reach the predetermined target attainment temperature.
When an embodiment of the present invention is applied, a method of estimating a load based on learning in order to effectively control cooling or heating of an air conditioner and an apparatus applying the same can be provided.
The effects of the present invention are not limited to the above-described effects, and those skilled in the art can easily derive various effects of the present invention from the structure of the present invention.
Drawings
Fig. 1 is a diagram illustrating an operation process of an air conditioner operating in a fast mode and a comfort mode.
Fig. 2 is a front view showing a structure of an indoor unit of an air conditioner according to an embodiment of the present invention.
Fig. 3 is a diagram showing the configuration of the control module 100 based on the internal learning according to the embodiment of the present invention.
Fig. 4 is a diagram showing the relationship between the cloud server 300 that executes external learning and the control module 200 based on external learning, and the respective components.
FIG. 5 is a diagram illustrating the process by which the in-house learning based control module calculates the result factor from the learning factor, in accordance with one embodiment of the present invention.
Fig. 6 is a diagram illustrating a process in which the external learning-based control module calculates a result factor from the learning factor according to an embodiment of the present invention.
FIG. 7 is a diagram illustrating the process by which the internal learning based control module operates in accordance with one embodiment of the present invention.
FIG. 8 is a diagram illustrating the operation of an external learning based control module in accordance with an embodiment of the present invention.
Fig. 9 is a diagram of the parameter generation unit calculating the input factor according to the embodiment of the present invention.
Fig. 10 is a diagram showing a configuration of a learning algorithm constituting the learning means in the embodiment of the present invention.
Fig. 11 is a diagram showing a change in the operation mode when the embodiment of the present invention is applied.
Fig. 12 is a diagram showing a procedure in which the operation mode control unit controls the operation mode according to the load level of an embodiment of the present invention.
Fig. 13 is a diagram showing the structure of the learning unit according to the embodiment of the present invention.
Fig. 14 is a diagram showing an exemplary configuration of a learning unit according to an embodiment of the present invention.
Fig. 15 is a diagram showing interaction between the supply air speed and the cooling air and parameters of an air conditioner for discharging the cooling air according to an embodiment of the present invention.
Fig. 16 is a view illustrating interaction between a supply air speed and heating air and parameters of an air conditioner for discharging heating air according to another embodiment of the present invention.
Wherein the reference numerals are as follows:
1: an indoor unit 2: outdoor machine
15: air blowing units 100, 200: control module
110. 210: parameter generation units 160, 360: learning unit
190. 290: operation mode control section 300: cloud server
Detailed Description
Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art to which the present invention pertains can easily carry out the embodiments. The present invention can be realized in various forms, and is not limited to the embodiments described herein.
The same or similar components are denoted by the same reference numerals throughout the specification, and portions not related to the present invention are omitted for the sake of accurately describing the present invention. Some embodiments of the invention are described in detail with reference to the exemplary drawings. When reference numerals are given to components in each drawing, the same reference numerals are used as much as possible for the same components in different drawings. In describing the present invention, when it is determined that specific descriptions of related well-known structures or functions may make the present invention unclear, detailed descriptions thereof may be omitted.
In describing the components of the present invention, terms such as first, second, A, B, (a), (b), and the like may be used. The above terms are only used to distinguish the above-mentioned components from other components, and the nature, order, sequence, number, and the like of the corresponding components are not limited by the above terms. When it is described that a certain constituent element is "connected", "coupled" or "in contact with" another constituent element, it is possible that the constituent element is directly connected or in contact with the other constituent element, but it should be understood that it is possible that another constituent element is "interposed" between the constituent elements or that the constituent elements are "connected", "coupled" or "in contact with" another constituent element.
In implementing the present invention, for convenience of description, the components may be detailed and described, and these components may be implemented in one device or module, or one component may be distributed among a plurality of devices or modules.
In this specification, components constituting an air conditioner are classified into an outdoor unit and an indoor unit. An air conditioning system is composed of more than one outdoor unit and more than one indoor unit. The relationship between the outdoor unit and the indoor unit may be 1:1, 1: N, or M: 1.
The present invention can be applied to all devices for controlling cooling or heating. However, for ease of illustration, refrigeration is described with emphasis. When applied to heating, embodiments of the present invention may be applied to a process of raising a temperature and a mechanism of maintaining the raised temperature.
Fig. 2 is a front view showing a structure of an indoor unit of an air conditioner according to an embodiment of the present invention.
The indoor unit of the air conditioner may be a buried type or a vertical type installed in a ceiling. Or may be a wall-mounted type provided on a wall, or may be in a movable form. Fig. 2 shows the indoor unit 1 in various embodiments, but the invention is not limited thereto. The indoor unit 1 can be connected to an outdoor unit 2 disposed in a separate space.
The air conditioner may be constituted by a floor type air conditioner installed standing on a floor in a room as an object of air conditioning, and in this case, the air conditioner further includes a base 20 placed on the floor in the room and supporting the air conditioning module 10.
The air conditioning module 10 may be disposed in such a manner as to be placed on the base 20, in which case the air conditioning module 10 may suck air at a prescribed height in the room for air conditioning.
The air conditioning module 10 may be detachably coupled to the base 20, or may be integrally formed.
The air conditioning module 10 can discharge air from the air blowing unit 15. The blowing unit 15 may be divided into right and left parts as 11 and 12. The air conditioning module 10 can collectively eject air on the front side. In addition, according to the embodiment, the air conditioning module 10 can discharge air from the air discharge ports arranged in a plurality of directions such as a side surface or an upper surface. The air blowing means 15 can control the air speed by the operation mode control means 190 and 290 described later. In one embodiment, the blowing unit 15 may discharge wind including a plurality of wind speeds, and for this purpose, one or more individual blowing fans may be controlled.
On the other hand, an intake unit for taking in indoor air may be disposed at the air conditioning module 10. The control module 100 that controls the indoor unit 1 without being recognized from the outside may be disposed in the indoor unit 1. For convenience of explanation, fig. 2 shows the control module as a dotted line disposed inside the indoor unit 1.
The outdoor unit 2 controls the temperature of the air (wind) discharged from the air blowing unit 15. In one embodiment, the compressor of the outdoor unit 2 may compress a gaseous refrigerant into a high temperature and high pressure state and discharge the compressed gaseous refrigerant to supply the refrigerant air to the indoor unit 1. In addition, when the heating function is provided, the outdoor unit 2 can supply heating air to the indoor units 1 using a predetermined heat pump. The manner in which the outdoor unit 2 supplies cooling or heating air to the indoor units 1 using a refrigerant or a heat pump may be provided in various ways, and the present invention is not limited thereto.
The indoor unit 1 exemplarily observed in fig. 2 operates to reach a set state after measuring the state of indoor air. However, in order to effectively perform the operation of the indoor unit in the process of reaching the specific state, it is necessary to reflect various elements before and after the specific state. Then, when the operation of the indoor unit is more precisely controlled by the learning model based on each element, an efficient operation can be realized.
Next, in the present specification, in the first embodiment, the control module 100 is configured in the indoor unit 1 to calculate various parameters and perform learning based on the calculated parameters. Then, the control module 100 that is arranged in the indoor unit 1 to be observed calculates a configuration suitable for the operation mode of the indoor unit based on the learning result. This configuration is referred to as an Internal Learning Based control module. In the internal learning, the air conditioner operates using the operation pattern information, which is the result factor calculated by inputting the parameter to the built-in learning unit 160, as an embodiment.
In addition, in the present specification, in the second embodiment, the control module 100 is configured in the indoor unit 1 to calculate various parameters and provide the calculated results to the cloud server. Then, the cloud server performs learning based on the parameters transmitted by the various indoor units. Then, the observation cloud server calculates a configuration suitable for the operation mode of the indoor unit based on the result of the learning. This configuration is referred to as an External Learning Based (External Learning Based) control module. Based on the external learning, as an embodiment, after the air conditioner transmits the parameters to an external cloud server, the air conditioner operates using the operation pattern information, which is the result factor transmitted from the cloud server.
In the present specification, the first section refers to a section that is operated at the first cooling capacity (the first heating capacity in the heating) in accordance with a set temperature (a target temperature set by a user). The second section is a section in which the vehicle is operated at a second cooling capacity (first heating capacity at the time of heating) different from the first cooling capacity after the section in which the vehicle is operated at the first cooling capacity. Therefore, when the user operates the air conditioner or the smart control at a specific time, the air conditioner is operated at a first cooling capacity (e.g., a maximum cooling capacity) corresponding to the set temperature. Which may be indicated as a fast run mode (or simply fast mode for ease of explanation). The fast operation mode can be operated only within a preset time range, after which the comfort operation mode is operated, and information required for a specific operation in the comfort operation mode is operation mode information.
The rapid operation mode may start operation when the air conditioner starts operation and the indoor temperature differs from the target set temperature by a predetermined magnitude or more. Or the air conditioner may be operated in response to an input signal indicating that the rapid operation mode is driven, when the air conditioner receives the input signal. For example, instead of separately controlling the temperature on the remote controller, the air conditioner performs the rapid operation mode in response to pressing a button for instructing the operation using artificial intelligence or for instructing the operation in a body-adaptive manner, and switches to the comfort operation mode described later without requiring the user to perform additional control.
Then, after the set temperature is reached or the maximum time allocated to the rapid mode has elapsed, the parameter generated during the operation in the rapid mode as described above is input as a learning factor, and the operation is performed at a second cooling capacity (for example, an overload, a standard load, or a small load determined by the magnitude of the output capacity of the cooling operation) different from the first cooling capacity. The above may be indicated as a comfort mode of operation (or simply comfort mode for ease of description). The second cooling capacity (overload, standard load, small load determined by the magnitude of the output capacity of the cooling operation) corresponds to the load of the space in which the air conditioner is installed, and is calculated as the operation mode information.
In one embodiment, the air flow or wind speed is increased during overload, and in one embodiment, the air flow or wind speed is decreased during light load. In one embodiment, the air flow or wind speed is maintained at standard load.
In the first cooling capacity, a cooling (or heating) function is provided to a user according to a maximum cooling (or heating) capacity, and then if a predetermined level (or time) is reached, energy less than the maximum capacity is used. Alternatively, in the first cooling capability, the air conditioner may be operated to achieve a comfortable state by using more energy. During operation at the second cooling capacity, the temperature may gradually increase. Since heating is the opposite of the above, the temperature may gradually decrease during operation at the second heating capacity.
In the first interval, when a user selects a specific function such as smart control or initially turns on the air conditioner or presses a button such as a power key, the fast operation is initially performed for a predetermined period of time and may be operated at a predetermined temperature later.
Fig. 3 is a diagram showing the configuration of the control module 100 based on the internal learning according to the embodiment of the present invention.
The parameter generation unit 110 generates parameters, such as temperature or humidity measured or sensed at the indoor unit 1, a rate of change in the temperature and humidity, time consumed for each change, and the like. The sensing unit 120 may sense a temperature or humidity required for the parameter generation unit 110 to generate the parameter. The sensed values are provided to the parameter generating unit 110, and the parameter generating unit 110 may accumulate the sensed values in an additional memory and then generate the parameters. Therefore, the parameter generation unit 110 can extract a learning factor for deriving a factor to be input to the learning unit 160 based on the environmental factor and the product control information recognized by the control module 100 configured at the indoor unit 1.
The target state storage unit 130 includes information such as a target temperature or a target humidity set in the indoor unit 1. The target state may include more than one temperature or more than one humidity. In addition, the target state storage unit 130 may also selectively include two or more temperatures or two or more humidities. In this case, which value is set as the target temperature or the target humidity of the indoor unit 1 currently in operation may be selected or set by the control of the central control unit 150.
The interface unit 140 can be used for the user to control the temperature, humidity, air volume or wind direction of the indoor unit 1, and provide interfaces such as button type, remote control type or remote control type. In addition, the interface unit 140 may receive an interrupt input for changing the air speed, the air volume, or the temperature of the air discharged from the blower unit 15. The interrupt input may be stored as prescribed information in the learning unit 160.
The learning unit 160 continuously accumulates the parameters (learning factors) generated by the parameter generation unit 110. The accumulated parameters may be applied to a deep learning structure inside the learning unit 160, and an optimum operation mode in which the indoor unit 1 can operate is calculated based on a change in temperature, humidity, or the like up to now. The operation modes may include various modes, and as an embodiment, three operation modes of a light load mode/a standard load mode/an overload mode may be included.
The information input as the learning factor of the learning unit 160 may be information generated or stored in the parameter generation unit 110, the target state storage unit 130, or the like. Further, the information input as the learning factor of the learning unit 160 may be information calculated or converted by the central control unit 150. The learning unit 160 may estimate the load level using a prescribed learning algorithm.
Alternatively, the learning unit 160 may set the corresponding load degree based on the state of operation so far. The operation mode may be set to-10%, -20%, etc.
The central control unit 150 controls each component and can finally calculate an operation method required when the indoor unit 1 operates. The operation method of the indoor unit 1 can be classified into various methods. For example, the load of the current indoor state is estimated by level and calculated as operation mode information such as overload mode/normal load mode/light load mode. The learning unit 160 calculates an operation mode required when the current indoor unit 1 is operated based on the change in temperature or humidity, the time, and the like, and thereby the central control unit 150 can control a specific load state.
The operation mode control unit 190 operates based on the operation mode calculated by the central control unit 150, and may be diversified according to the type of the calculated mode. When the operation mode is composed of three kinds of the above-described light load mode/standard load mode/overload mode, the central control unit 150 may select the operation mode among the three operation modes as the operation mode selected based on the above-described parameter generation unit 110 and the learning unit 160.
The operation mode control unit 190 can control the blower unit 15 and the outdoor unit 2 based on the operation mode described above. For example, control may be performed to control the speed of the air discharged from the air blowing unit 15, the amount of the refrigerant gas compressed by the compressor constituting the outdoor unit 2 and discharged, or the like.
In one embodiment, when the operation mode set by operation mode control section 190 corresponds to the overload mode, one or more of the air speed and the air volume of air blowing unit 15 are increased, and one or more of the air speed and the air volume of air blowing unit 15 can be decreased in the light load mode. In the standard load mode, the air blowing unit 15 may maintain the wind speed. Similarly, when the operation mode set by the operation mode control unit 190 corresponds to an overload, the outdoor unit 2 may control the internal compressor so that the compressor operates and operates at the maximum output. In addition, the compressor may be powered OFF (OFF) in the case of a small load.
That is, the operation mode control unit 190 may control the air blowing unit 15 or the outdoor unit 2 based on the operation mode information in which the comfortable operation mode is set after the section in which the rapid operation mode is operated. Based on the operation mode information, the operation mode control unit 190 can control the wind direction and the wind volume of the blower unit 15.
As described above, the rapid operation mode is operated only within a predetermined time range (e.g., a preset time range such as twenty minutes or thirty minutes) to rapidly change the indoor temperature. Then, the central control unit 150 instructs the operation mode control unit 190 to operate in the comfort operation mode using the operation mode information calculated based on the parameter calculated in the fast operation mode.
The operation mode control unit 190 may control the blowing unit 15 and the outdoor unit 2 by various methods. In the embodiment in which the operation modes are subdivided, various operation modes may be implemented by separately controlling the wind speed and the on/off of the compressor of the outdoor unit.
The structure of the air conditioner including the control module 100 based on the internal learning is summarized as follows. This air conditioner includes: an air blowing unit 15 for blowing air supplied from the outdoor unit to cool or heat; a parameter generation unit 110 that calculates one or more parameters in the first section; a learning unit 160 that receives the parameter as a learning factor and outputs operation mode information indicating an operation mode in a second section subsequent to the first section; an operation mode control unit 190 for controlling the air blowing unit 15 or the outdoor unit 2 in the second section based on the operation mode information; and a central control unit 150 for controlling the parameter generation unit 110, the learning unit 160, and the operation mode control unit 190, and the air conditioning mechanism causes the power consumption per hour of the second section to be smaller than the power consumption per hour of the first section. The air conditioner operates in a fast mode in a first section, and the air conditioners 1, 2 may operate based on less power consumption than the fast mode in the first section in a comfort mode which is a second section after the fast mode, using a learning factor calculated during the operation in the fast mode.
In an embodiment of the calculable parameter, the parameter may be at least one of an indoor initial temperature at a start time of the first section (a section operated in the first cooling capacity, for example, the fast mode), a target set temperature of the first section, a temperature change rate of a preset initial section of the first section, a temperature change rate of the first section, and a time interval between the start time and the end time of the first section. Here, in order to calculate each parameter, information calculated by the sensing unit 120 or the target state storage unit 130 may be utilized.
On the other hand, if the air conditioner receives an interrupt input from the interface unit 140 during the second interval (the interval operating in the second cooling capacity, e.g., the comfort mode), the learning unit 160 may be updated based on the interrupt input.
In more detail, the central control unit 150 may provide the operation mode information and the interrupt input to the learning unit 160 to update the learning unit 160 or update the operation mode information. Then, when the operation mode information is changed or updated due to the update of the learning unit 160, the updated operation mode information may be provided to the operation mode control unit 190 again.
Fig. 4 shows the structure of the external learning-based control module according to an embodiment of the present invention. Fig. 4 is a diagram showing the relationship between the cloud server 300 that executes external learning and the control module 200 based on external learning, and the respective components.
Since the parameter generation unit 210, the sensing unit 220, the target state storage unit 230, the interface unit 240, and the operation mode control unit 290, which are components of the control module 200 based on external learning, have the same configurations as the parameter generation unit 110, the sensing unit 120, the target state storage unit 130, the interface unit 140, and the operation mode control unit 190 described in fig. 3, the description of fig. 3 may be used instead.
The central control unit 250 controls each component and finally controls the communication unit 280 to transmit the learning factor, which is a parameter required for calculating an operation method required for the operation of the indoor unit 1, to the cloud server 300. The server control unit 350 of the cloud server 300 receives the learning factor transmitted from the control module 200 from the communication unit 380, inputs the factor to the learning unit 360, and calculates an operation mode suitable for the corresponding control module 200. The information of the calculated operation mode is transmitted to the control module 200 through the communication unit 380.
As shown in fig. 10, 13, and 14, the learning unit 360 includes an artificial neural network, and may generate links (link), biases (bias), or weights (weight) of each link, which are configured at a hidden layer and each input/output factor constituting the artificial neural network, in a learning process, and store information updated from the outside. In this case, the learning unit 360 may store the different versions to the cloud server 300.
The cloud server 300 receives the learning factor from the plurality of control modules, and may calculate an operation mode corresponding to the learning factor. In addition, the learning factors provided by the plurality of control modules may be continuously input to the learning unit 360 to update the learning unit 360. The learning unit 360 may estimate the load level using a prescribed learning algorithm.
The cloud server 300 of fig. 4 is summarized as follows.
The communication unit 380 receives one or more parameters calculated in the operation mode of the first section from the M air conditioners, and transmits operation mode information corresponding thereto to the M air conditioners, respectively. Then, the learning unit 360 inputs the parameter of the first air conditioner of the M air conditioners as a learning factor, and outputs operation mode information indicating the operation mode of the first air conditioner on a second section subsequent to the first section.
The first cooling capacity in the first zone and the second cooling capacity in the second zone may be set to be different from each other. For example, the first cooling capacity may be an operation mode in which the air speed and the air volume, the temperature of the refrigerant, and the like are increased by the maximum cooling capacity to rapidly decrease the temperature. This is an environment for providing a user with a temperature that quickly feels comfortable for a predetermined time (short time). On the other hand, the second cooling capacity means that less energy is consumed while maintaining a temperature after a temperature environment of a comfortable level is provided first. Alternatively, it is also possible to provide the second cooling capacity with an increased wind speed or wind volume without reaching a temperature at which a comfortable level is felt.
The server control unit 350 may control the learning unit 360, the communication unit 380. Further, the operation mode information may be output in which the power consumption amount per hour in the second section of the first air conditioner is smaller than the power consumption amount per hour in the first section of the first air conditioner. The output operation mode information is transmitted to the corresponding air conditioner (first air conditioner) via the communication unit 380.
As an example of the parameter transmitted by each air conditioner, information calculated by each air conditioner is included. In one embodiment, the parameter calculated in the first section may be at least one of an indoor initial temperature at a start time of the first section, a target set temperature of the first section, a temperature change rate of a preset initial section in the first section, the temperature change rate of the first section, and a time interval between the start time and the end time of the first section. For calculating each parameter, information calculated in the sensing unit 120 or the target state storage unit 130 may be utilized.
If the embodiment shown in fig. 3 or 4 is applied, it is possible to estimate the load of the current indoor state by level (e.g., small load, standard load, overload) by learning the environmental factor and the control information until the time when the target temperature is reached after the air conditioner is started.
Further, it is possible to learn the correlation between the learning factor at the time when the target temperature is reached and the temperature pattern of the subsequent cooling (or heating) and automatically operate the customized cooling mode (or the customized heating mode) according to the set load, and it is possible to automatically operate the power-saving and comfort cooling (or heating) mode after the target temperature is reached according to the indoor load level without the need for the user to separately operate the air conditioner.
In addition, if each air conditioner transmits an interrupt input occurring during the second interval operation, the communication unit 380 may receive such interrupt input from each air conditioner and update the learning unit 360 based on the interrupt input.
In more detail, the server control unit 350 may provide the operation mode information and the interrupt input to the learning unit 360 to update the learning unit 360 or update the operation mode information. Then, when the operation mode information is changed or updated due to the update of the learning unit 360, the updated operation mode information may be provided to the air conditioner through the communication unit 380 again.
In the case of applying the embodiment shown in fig. 3 or 4, the load level is estimated to reflect the size of the space, the insulation state, the temperature difference between the indoor and outdoor, etc., which affect the cooling efficiency, after the air conditioner reaches the target temperature, so that the air conditioner can effectively cool after reaching the target temperature. This is a result calculated by the learning units 160, 360 receiving various learning factors obtained in the process of reaching the target temperature. The learning factor calculated during the process of reaching the target temperature will be observed.
FIG. 5 is a diagram illustrating the process by which the in-house learning based control module calculates the result factor from the learning factor, in accordance with one embodiment of the present invention.
The parameter generation unit 110 calculates a parameter, i.e., a learning factor, so that the control module 100 based on the internal learning can calculate a result indicating an operation method, i.e., an operation mode, required when the indoor unit is operated. In an embodiment of the parameters generated by the parameter generation unit 110, the initial indoor temperature sensed when the air conditioner starts to start, the target set temperature, the initial N-minute temperature change rate, the fast interval temperature change rate, the time to reach the target temperature, and the like may be used. The initial N minute temperature change rate refers to a rate of temperature change during, for example, the initial three minutes or the initial five minutes. Of course, this may also utilize the magnitude of the temperature change.
The foregoing parameters are exemplary, and in one embodiment, the parameters are the time when the indoor unit 1 performs the operation or various information calculated in the process.
The parameters generated by the parameter generation unit 110 are input to the control module 100, and the learning unit 160 of the control module 100 performs predetermined learning using the input learning factor. The learning unit 160 may apply a deep learning (deep learning) module. Then, learning section 160 inputs the input parameters into the network constituting the deep learning, and calculates a predetermined result factor, i.e., an operation pattern. One example of the parameters is an initial temperature TempInit, a target set temperature TempTarget, an initial three-minute temperature change rate InitRate, a fast interval temperature change rate PowerRate, and a time to reach the target temperature PowerTime. The calculated operation mode may increase or decrease the load, such as overload/normal load/light load, based on the current operation mode. Alternatively, the calculated operation mode may be set numerically for load adjustment based on the current operation mode.
The operation mode control unit 190 controls the indoor unit 1 or the outdoor unit 2 according to the calculated operation mode.
Fig. 6 is a diagram illustrating a process in which the external learning-based control module calculates a result factor from the learning factor according to an embodiment of the present invention. Fig. 6 shows a process in which the plurality of indoor units 1 and the learning unit 360 of the control module and cloud server 300 calculate information of an operation mode required for operating each indoor unit 1.
The parameter generation unit 210a constituting the control module of the first indoor unit 1a calculates one or more parameters, i.e., learning factors, so that it can calculate a result indicating an operation method, i.e., an operation mode, necessary for operating the indoor unit. As shown in fig. 5, an example of the parameter generated by the parameter generating unit 210a may be an indoor initial temperature sensed when the air conditioner starts up, a target set temperature, an initial N-minute temperature change rate, a fast section temperature change rate, a time to reach the target temperature, and the like. The initial N minute temperature change rate refers to, for example, a rate of temperature change at the initial three minutes or the initial five minutes. Of course, this may also use the magnitude of the temperature change.
The calculated parameters are transmitted to the cloud server 300 as in S31a, and the cloud server 300 inputs the received parameters to the learning unit 360.
Similarly, the parameter generation unit 210b constituting the control module of the second indoor unit 1b also calculates a parameter, i.e., a learning factor, so that it can calculate a result indicating an operation method, i.e., an operation mode, necessary for operating the indoor unit. Then, the calculated parameters are transmitted to the cloud server 300 as in S31b, and the cloud server 300 inputs the received parameters to the learning unit 360.
The learning unit 360 in the cloud server performs prescribed learning using the input learning factor. The learning unit 360 may apply a deep learning (deep learning) module. Then, learning section 360 inputs the input parameters to a network constituting deep learning, and calculates a predetermined result factor, i.e., an operation pattern. This can be calculated for each indoor unit 1a, 1 b. The calculated operating mode may then indicate an increase or decrease in load, such as overload/standard load/light load, with respect to the current operating mode. This means that the air conditioner performs cooling (or heating) output corresponding to different loads (overload/standard load/small load). Alternatively, the calculated operation mode may be set numerically for load adjustment based on the current operation mode.
The individually calculated operation mode of each indoor unit 1a, 1b is transmitted to each indoor unit 1a, 1b again (S32a, S32b), and the operation mode control unit 290a, 290b of each indoor unit is operated according to the operation mode transmitted from the cloud server 300. That is, the operation mode control units 290a and 290b control the indoor units 1a and 1b or the outdoor unit 2 according to the calculated operation mode.
As shown in fig. 5 and 6, the learning result can be derived by applying the environmental factor and the product control information for load estimation to the learning input factor, and inputting the learning factor to the learning unit 160 disposed in the indoor unit 1 or the learning unit 360 disposed in the cloud server 300. The result may then be indicated as the operational mode. In an embodiment, the load may be indicated by level (overload/standard load/light load). Alternatively, the load may be set +/-based on the current operation, or the adjustment of the load may be indicated as a percentage%.
Based on the load estimation method of the internal learning based control module as shown in fig. 5, the environmental factor and the product control information may be generated as parameters called learning factors and provided to the learning unit 160 as the learning factors, and the load result is derived by applying a learning algorithm configured in the learning unit 160.
Based on the load estimation method of the external learning based control module as shown in fig. 6, an environment factor and product control information may be generated as parameters called learning factors and provided to the learning unit 360 of the cloud server 300 as the learning factors, and a load result is derived by applying a learning algorithm configured in the learning unit 360.
In the embodiment shown in fig. 6, after each air conditioner starts the intelligent control to operate at the maximum cooling capacity, enters the comfort mode or after a predetermined time elapses, a prescribed parameter may be transmitted to the cloud server 300 to receive operation mode information for setting a cooling load to be performed in the comfort mode. The intelligent control means that the air conditioner automatically operates a fast mode and a comfort mode without user control.
FIG. 7 is a diagram illustrating the process by which the internal learning based control module operates in accordance with one embodiment of the present invention.
In the air conditioner driven based on the learning, the air blowing unit 15 discharges air in the first section and is used for cooling or heating of the air conditioner (S40). In one embodiment, the first interval is the fast mode and the second interval immediately following the first interval is the comfort mode. The first interval includes reaching the target attainment temperature for a short time. The second section is a section capable of maintaining the indoor temperature within a predetermined error range with reference to the target temperature.
The parameter generation unit 110 calculates a learning factor, which is an input factor to be input to the learning unit 160 (S41). That is, the parameter generation unit 110 may calculate various parameters in the first section. Then, when the input factor is supplied to the learning unit 160(S42), the learning unit 160 estimates the load according to the input factor (S43). In one embodiment, the learning unit 160 receives the calculated parameter as a learning factor and outputs operation mode information indicating an operation mode of a second section subsequent to the first section.
Thereafter, when the central control unit 150 provides the estimated load level (the outputted operation mode information) to the operation mode control unit 190(S44), the operation mode control unit 190 controls the operation mode of the indoor unit 1 or the outdoor unit 2 (S45).
When the central control unit 150 provides the operation mode information output from the learning unit 160 to the operation mode control unit 190 while observing S44 and S45, the operation mode control unit 190 controls the air blowing unit 15 or the outdoor unit 2 in the second zone based on the operation mode information as an example. In the case based on the procedure of fig. 7, if the power consumption amount per hour of the first section and the power consumption amount per hour of the second section of the air conditioner are compared, the power consumption amount per hour of the second section is smaller than the power consumption amount per hour of the first section. That is, after the fast mode is operated in the first section so that the indoor temperature reaches the target attainment temperature in a short time, the target attainment temperature is maintained with a small amount of power consumption after the target attainment temperature, or the indoor temperature is maintained within a predetermined range. The operation mode information of the second section is determined by the parameter calculated by the learning unit 160 in the first section. Therefore, when the parameters calculated in the first section are different, the operation mode information in the second section may be different.
FIG. 8 is a diagram illustrating the operation of an external learning based control module in accordance with an embodiment of the present invention.
The parameter generation unit 210 extracts an input factor to be input to the learning unit 360 of the cloud server 300, that is, a learning factor (S51). Then, when the input factor is transmitted to the cloud server 300(S52), the learning unit 360 of the cloud server 300 estimates the load (S53). At this time, since the cloud server 300 receives the learning factor from a plurality of products, the learning factor may be input to the learning unit 360 separately for each product to estimate the load of each product. Thereafter, the cloud server 300 transmits the estimated load level to the corresponding product (S54). The central control unit 250 of each product provides the received load level to the operation mode control unit 290, and the operation mode control unit 290 of each product controls the operation mode of the indoor unit 1 or the outdoor unit 2 (S55).
The process of fig. 8 is summarized as follows.
The cloud server controls driving of the plurality of air conditioners based on the learning. In each air conditioner, each parameter generation unit 210 calculates a parameter in the operation mode of the first section (S51). Then, the communication unit 380 receives one or more parameters calculated in the operation mode of the first section from the first of the plurality of air conditioners (S52). The process can be continuously accumulated, parameters calculated by a plurality of air conditioners can be accumulated in the cloud server 300, and the cloud server 300 can be additionally provided with a database.
The learning unit 360 inputs the parameter received from each air conditioner as a learning factor, and outputs operation mode information indicating an operation mode of the corresponding air conditioner on a second section subsequent to the first section (S53). Then, the communication unit 380 may transmit the operation mode information output by the learning unit 360 to each air conditioner according to the control of the server control unit 350.
In fig. 8, the cloud server 300 provides operation mode information required for the corresponding air conditioner to operate in the second section based on the parameter transmitted by each air conditioner. In this case, the operation mode information in the second section is characterized in that the power consumption of the air conditioner in the second section per hour is smaller than the power consumption of the air conditioner in the first section per hour.
In fig. 8, the learning unit of the cloud server may calculate various information required for estimating the load in a learning process or directly receive the information from the outside. For example, as shown in fig. 10, fig. 13, and the like, the learning unit of the cloud server may be constituted by hidden layers, and information on the setting of each node of these hidden layers, the link or bias between nodes, the weight of the link or node, and the like may be changed in the learning process after being set in advance. In addition, information separately monitored from the outside or calculated from learning may be applied to the learning unit. Here, the external means a server independent from the cloud server. Alternatively, an additional storage medium such as a memory card may be combined with the cloud server to apply the information stored in the storage medium to the learning unit.
Based on the embodiments of fig. 7 or 8, a learning-based load estimation method may be applied to effectively control cooling of the air conditioner. In this process, the load estimation (output of the operation mode information) in the second section may be estimated by level by learning the environmental factor calculated at the time of reaching the target temperature, the control information, and the temperature mode that changes according to the cooling after the time of reaching the target temperature.
In addition, the temperature is set according to the estimated load level so that power saving or comfortable cooling is possible in the second section, and the learning units 160 and 360 may output operation mode information for changing the air volume/wind direction and the like.
As shown in fig. 3 and 7, the control module 100 based on the internal learning is provided in the air conditioner as an example in which the learning logic (learning means 160) for load estimation is mounted on a product. In addition, when the learning logic (learning unit 360) for load estimation is mounted on the cloud server, the control module 200 based on external learning is provided to the air conditioner, and the cloud server 300 may perform analysis or relearning after receiving the environmental factors and the product control information from the control module 200 through wireless communication.
Fig. 9 is a diagram of the parameter generation unit calculating the input factor according to the embodiment of the present invention. Temperature is shown on the Y-axis and time is shown on the X-axis, indicating the change in temperature with time. Fig. 9 shows an example of input factors for learning, and an example of calculation of learning factors such as temperature change rates derived from changes in control information and temperature on the product side with time.
In the above-described embodiment, as input factors that the parameter generation unit can generate, the indoor initial temperature, the target set temperature, the initial N-minute temperature change rate, the rapid interval temperature change rate, the time to reach the target temperature, and the like are observed. Here, N may be selected in various ways, three minutes in the embodiment of fig. 9.
Fig. 9 shows that the parameter generation units 110, 210 can calculate the initial temperature TempInit and the target set temperature TempTarget. The initial temperature TempInit may be calculated by sensing the initial temperature in the room using the sensing units 120, 220. The target set temperature TempTarget is calculated based on the target set temperature stored in the target state storage unit 130, 230. On the other hand, the parameter generation units 120, 220 may calculate the initial three-minute temperature change rate InitRate with a/b.
b refers to the time elapsed after the air conditioner is operated. For example, it may be three minutes or five minutes. a refers to the magnitude of the temperature change from TempInit over the b period.
On the other hand, the fast interval temperature change rate PowerRate can also be calculated by c/d. c refers to the temperature difference between TempTarget and TempInit. Thus, c may be "TempInit-TempTarget". Then, the PowerTime until the target temperature is reached is calculated by d. In one embodiment, d represents the time required to perform operation at maximum cooling capacity during at most M minutes to reach the set temperature at the fastest rate, including a time value of fifteen or twenty minutes in one embodiment. In addition, the cooling capacity to be performed in the comfort mode (second interval) may be set at time d to determine overload/standard load/small load.
The first section (the rapid operation mode or the rapid section) refers to a mode in which the air conditioner is operated after the initial driving to reach the target temperature. In one embodiment, a high-speed cooling operation mode in which the maximum cooling capacity of the air conditioner reaches the target set temperature at the time of initial cooling is set as an embodiment of the first section mode.
In fig. 9, the target temperature may be set to a specific target temperature value, but may be a temperature within a predetermined range. For example, when the target temperature value is 20 degrees, as an example of reaching the target temperature, it may be that the current temperature reaches 20 degrees. However, as another example of the target temperature being reached, even if the current temperature reaches a state of +1 degree or-1 degree (i.e., 19 degrees to 21 degrees) with 20 degrees as a reference, it may be determined that the target temperature is reached.
This can be applied to a case where a time range in which the fast interval can be operated is set in advance. For example, assume a case where a time for which the maximum cooling capacity is operated in the fast zone (fast-possible time) is preset to, for example, ten minutes or fifteen minutes. If the air conditioner is operated for a time that is faster than the target temperature after the start of the operation but does not reach the target temperature, the parameter generation unit 110, 210 may include the currently reached temperature in the learning factor instead of the target temperature.
Then, when the target temperature is reached, it is possible to change from the first interval (fast interval) to the second interval (comfort operation mode or comfort interval). In addition, even if the target temperature is not reached, the operation mode of the air conditioner may be changed to the comfort zone when a predetermined time elapses or the target temperature is approached. The second interval (comfort interval) is comfort operation as an example, and includes maintaining the set temperature after the target temperature is reached and operating in an automatic mode (intermittent wind).
When the influence of the outside on the space is large or the space is wide, the indoor temperature cannot reach the target temperature. Therefore, even when the temperature is close to the target temperature to some extent, the mode can be changed from the fast interval (first interval) to the comfortable interval (second interval).
Thereafter, an operation mode in a different manner from the first interval (fast interval) may be selected in the second interval (comfort interval). As described above, the learning units 160 and 360 may calculate the operation mode of the air conditioner in the comfort zone, i.e., the operation mode, using the five learning factors (TempInit, TempTarget, InitRate, PowerRate, PowerTime) generated by the parameter generation units 110 and 210.
In fig. 9, the absolute value of the temperature change rate per hour in the first interval operating in the fast operation mode is larger than the absolute value of the temperature change rate per hour in the second interval operating in the comfort operation mode. This is due to the rapid temperature change in the fast operation mode, and the maintenance of the changed temperature in the comfort operation mode.
In addition, in the second section (comfort operation mode), if the difference between the temperature of the space and the target set temperature is included in a predetermined temperature range set in advance, or such a difference is maintained for a predetermined time, the central control unit 150, 250 may instruct the operation mode control unit 190, 290 to switch to the rapid operation mode for rapidly cooling or heating the space. In this case, the operation mode control unit 190, 290 may control the blowing unit 15 and the outdoor unit 2 such that the air conditioner operates in a rapid operation mode for rapidly reducing the temperature of the space as shown in the first section. When the temperature difference is 2 degrees or more, or such temperature difference is maintained for five minutes or more, in order to rapidly reduce the indoor temperature, it is possible to operate in a rapid operation mode.
In addition, if the humidity is above a predetermined reference (e.g., 70% or more) or such humidity remains for five minutes or more in the comfort operating mode, it is possible to switch to the humidity control operating mode. The central control unit 150, 250 continuously monitors the humidity and, when a predetermined reference is reached, interrupts the comfort mode of operation and instructs the mode of operation control unit 190, 290 to begin the humidity control mode of operation. Thereafter, when the humidity reaches below a predetermined level (e.g., below 50%) and this state is maintained for a predetermined time or more (e.g., five minutes or more), the humidity control operation mode may be cancelled and the comfort operation mode may be started again.
According to the embodiment shown in fig. 9, the load may be estimated based on the environmental factor that the air conditioner can calculate. This includes calculating characteristics that the cooling environment of the space where the air conditioner is disposed has as various learning factors, and calculating the load based on the learning factors. In addition, the calculation of the load is not based on a simple function, but may use a deep learning algorithm provided by the learning unit 360 of the cloud server 300 or the learning unit 160 in the control module 100 of the air conditioner to calculate an optimal load corresponding to the learning factor. As a result, the air conditioner can select the power saving or the comfort cooling based on the load level to operate in the second section.
In the conventional art, since the environmental change is not considered after the fast interval, there is a problem in that the user controls the temperature again in the comfort interval after the end of the fast interval. However, in the embodiment of the present invention, the cooling state can be maintained in the comfort zone state based on the initial learning and the continuous learning without the user's additional control of the temperature.
In particular, in order for the user not to need additional air conditioning adjustment and to bring comfort to the user, cooling (or heating) is performed quickly at the beginning until reaching the vicinity of the target temperature (first section). Then, if the temperature reaches the vicinity of the target temperature, the air conditioner may determine whether to continue to maintain the operation mode of the first section, or to consume more electric power or less electric power than the operation mode of the first section, while maintaining cooling or heating. As an embodiment, the load determination is performed.
The values input to the learning units 160 and 260 as values for load determination include, as parameters, the time or temperature change, the initial value, the result value, or the magnitudes thereof calculated in the first interval so far.
In addition, the central control unit 150, 250 or the server control unit 350 of the cloud server 300 may sense a condition in which the air conditioner changes temperature during operation according to the operation mode calculated in the learning process. In this case, the nodes or links constituting the hidden layer of the deep learning of the learning units 160, 360 may be reconfigured or the weights may be changed to calculate a more suitable operation mode.
On the other hand, in fig. 9, when the humidity sensed at the end time of the first section is equal to or higher than the set reference, the central control units 150 and 250 may instruct the operation mode control units 190 and 290 to perform the humidity control operation mode. For example, if the humidity is 60% or more, the dehumidification function may be added according to the humidity control operation mode instead of the comfort operation mode.
When the humidity reaches a set reference or less (e.g., 50% or less) and this state is maintained for a predetermined time or more (e.g., five minutes or more), the control unit may cancel the humidity control operation mode and enter the comfort operation mode.
Fig. 10 is a diagram showing a configuration of a learning algorithm constituting the learning means in the embodiment of the present invention. As a design example of the learning structure, an example of a node-based structure showing an input factor, and a hidden layer (hidden layer) and an output factor constituting the learning structure is given.
The factors generated by the parameter generation units 110 and 210 are Input to the Input layers (Input layers) of the learning units 160 and 260. Five factors are provided, but various factors may be applied according to embodiments.
The learning units 160, 260 are configured with a plurality of hidden layers so that the correlation of the input factor with the edge of each layer can be calculated. For example, in fig. 10, three Hidden layers (Hidden Layer1, Hidden Layer2, and Hidden Layer3) are arranged.
Each node of the input layer has, as an example, five input values. The values input from the five input nodes are converted at the input level or may be output without conversion.
Then, values output from these input layers selectively form input values of twelve nodes in the first Hidden layer (Hidden layer 1). Likewise, the first Hidden layer (Hidden layer 1) applies the weight of the link to the input value, and calculates an output value according to the logic of each node.
The values output from the first Hidden layer selectively reform input values of eight nodes of the second Hidden layer (Hidden layer 2). Likewise, the second Hidden layer (Hidden layer 2) applies the weight of the link to the input value, and calculates an output value according to the logic of each node.
The values output from the second Hidden layer selectively reform input values of four nodes of a third Hidden layer (Hidden layer 3). Likewise, the third Hidden layer (Hidden layer3) applies the weight of the link to the input value, and calculates an output value according to the logic of each node.
Finally, the Output node (Output) can calculate the load level with three nodes. In the case of overload/normal load/light load as described above, each value may be input to the output node labeled Over, the output node labeled Medium, and the output node labeled Under, respectively.
In an embodiment, the operation mode may be indicated as overloaded if the output node marked Over is 1, the output node marked Medium and the output node marked Under are both 0. On the other hand, if the output node labeled Over is 1, the output node labeled Under is 1, and the output node labeled Medium is 0, the operation mode may be indicated as a standard load.
In another embodiment, the operation mode may be indicated as light load if the output node marked Over is 0, the output node marked Medium is 0 and the output node marked Under is 1. On the other hand, if the output node marked Over is 0, the output node marked Under is 0 and the output node marked Medium is 0, the operation mode may be indicated as standard load.
In fig. 10, links between nodes of each layer are not shown. During the learning process in which the learning units 160, 260 adjust the algorithm or the constituent elements of the algorithm, links may be newly created or added, or the weight assigned to each link may be changed. In addition, the number of hidden layers may be increased or decreased, or the number of nodes constituting a hidden layer may be increased or decreased.
The hidden layer and the input factor of fig. 10 initially collect data used in reality (live) by a plurality of actual households and laboratory (experimental) data for standard environmental tests to extract input/result factors, and may be learned in advance based thereon to set initial weight values of the hidden layer. Thereafter, while actual usage data is continuously collected by the cloud server, or the control module 100 in the same indoor unit, relearning of the entire data is applied to periodically perform weight update for each node, each link of the hidden layer.
In one embodiment, the DB collected for a predetermined time is used for the judgment of the output factor when the air conditioner is operated, so that a clustering technique (unsupervised learning: k-means algorithm) can be utilized in the load reference classification.
In addition, each hidden layer can directly use a general deep learning method. However, in another embodiment, the structure of each hidden layer used for learning may be changed or the weight may be updated. For example, the cloud server 300 may perform analysis after DB-converting information provided by a plurality of indoor units, so as to change the structures of a plurality of layers, nodes, and links in fig. 9. In addition, after the target temperature is set, the user's temperature is sensed for a predetermined time in a state of the comfort zone and the sensed temperature may be reflected. In fig. 13 described later, the user's adjustment of the wind speed, the air volume, the temperature, or the like may be included as an Interrupt input (Interrupt input) at an input node of the learning units 160, 360.
Fig. 11 is a diagram showing a change in the operation mode when the embodiment of the present invention is applied.
As shown in fig. 9 described above, when the air conditioner starts to operate, the parameter generation units 110, 210 continuously sense the temperature to calculate the graph shown in fig. 11, and in the process, the learning factor may be calculated. Three different calculation graphs are shown in fig. 11. First, a graph (indicated as G-Over) in which the operation mode is calculated as the overload is a graph indicated by a two-dot chain line, a graph (indicated as G-Medium) in which the operation mode is calculated as the standard load is a graph indicated by a solid line, and a graph (indicated as G-Under) in which the operation mode is calculated as the small load is a graph indicated by a one-dot chain line.
The first interval is an interval in which the compressor rapidly decreases the temperature in a short time with the maximum output or a large amount of output. The fast interval is taken as an example. Then, the target temperature TempTarget is reached according to the space condition, or the time when the target temperature is reached by the predetermined temperature difference is PowerTime. Until this time, the parameter generation units 110, 210 generate various parameters and supply the parameters to the learning units 160, 360. The learning units 160, 360 can calculate the operation mode using a plurality of parameters provided at PowerTime. The learning unit 160, 360 may analyze the change pattern of the cooling according to the first section at the time period indicated by 61. Then, the air conditioner may select an operation mode that provides a user with comfort while reducing power consumption in the second section.
When the overload operation mode is calculated by the learning units 160 and 360 according to the change mode that does not sufficiently reach the target temperature in the first section, or the time taken to reach the temperature is long, or the change rate of the initial N minutes is reflected, the air conditioner may be operated in the overload mode as shown in the graph indicated by G-Over.
When the standard load operation mode is calculated by the learning units 160, 360 to maintain the current temperature or the target temperature according to the change pattern sufficiently reaching the target temperature in the first section, or the time consumed to reach the temperature corresponds to the reference value, or the change rate of the initial N minutes is reflected, etc., the air conditioner may be operated in the standard load mode as shown in the graph indicated by G-Medium.
When the low load operation mode is calculated by the learning units 160 and 360 according to whether the temperature reached in the first section by the change mode is lower than the target temperature, or the time taken to reach the temperature is short, or the change rate of the initial N minutes is reflected, the air conditioner may be operated in the low load mode as shown in the graph indicated by the G-Under.
Even if the target temperature is the same, the air conditioner may be operated in different manners if the change pattern until the target temperature is reached is different. That is, according to the difference of the plurality of learning factors shown, after the target temperature is reached, the operation mode of performing further cooling or reducing cooling, etc. can be appropriately performed according to the condition of the indoor environment based on the different result analyzed along with the temperature change pattern of cooling.
In the conventional art, cooling (or heating) is performed in the first section in the same manner due to different indoor environmental conditions, and then, weak cooling (or weak heating) or excessive cooling (or excessive heating) occurs depending on the space. However, if the embodiment of the present invention is applied, cooling (or heating) may be appropriately performed in the second section according to the indoor environmental conditions, and thus a comfortable indoor environment may be maintained while reducing energy consumption.
Fig. 12 is a diagram showing a procedure in which the operation mode control unit controls the operation mode according to the load level of an embodiment of the present invention. In one embodiment, the observation is calculated to be a light load case. The air conditioner is operated in a fast mode to reach a target temperature from an initial temperature TempInit. In fig. 12, since the time zone represented by the first interval is a state before the target set temperature is reached, the air conditioner may be operated in the fast mode, in which case it may be operated in the cooling power mode using the maximum cooling power of the air conditioner. In this process, the air conditioner may sense a human body or sense a region of a space to more effectively reduce the temperature in the space.
On the other hand, since the time zone indicated by the second section is a state after the target set temperature is reached, the air conditioner may be operated in the comfort mode. The learning unit 160 of the control module or the learning unit 360 of the cloud server according to an embodiment of the present invention determines the operation mode at time t1 using the learning factors generated by the parameter generation units 110 and 210. That is, the load required for the air conditioner to operate in the second zone is determined. As a result of the determination, fig. 12 determines that the load is small to operate. This is because the operation mode control unit 190, 290 can control the indoor unit, the outdoor unit, and the like to reduce the human body adaptation time, increase the temperature, and reduce the air volume in the second section after the fast mode in the first section.
According to the small load control of the operation mode control units 190 and 290, at time t2, the temperature rises partially from the state (Temp-a) of the target temperature at time t1, for example, Temp-aa, and at time t3, the temperature rises again, for example, Temp-ab, and the temperature can be maintained at Temp-ab. The automatic setting of the target temperature to the temperature mainly used by the user is an example, but may be a temperature that the user set at most in the last N times (for example, 20 times). Alternatively, the external server may set a temperature that corresponds to the current temperature and is preferable based on the big data as the target temperature.
At t1, the cooling capacity after t1 may be set based on various parameters calculated until the target temperature is reached. For example, whether the current state is to provide an overloaded cooling capacity, a standard load cooling capacity, or a small load cooling capacity may be set based on previously learned information and calculation parameters. Then, the cooling capacity is gradually adjusted at the stages t2 and t3, and the operation is performed in the power saving operation. In this process, the target humidity may be matched according to an additional humidity control process when the humidity is high. Further, the time (t2, t3) at which the temperature is raised according to the user pattern, and the like can be calculated.
Although not shown in the drawings, when it is determined as a standard load, the human body adaptation time is reflected in a standard condition, so that it is possible to increase the temperature and switch the air volume to a weak air. In addition, when it is judged that the overload is generated, the air volume may be increased without increasing the temperature according to the human body adaptive condition.
According to an embodiment, the overload/standard load/light load may be determined as the operation type of the comfort operation mode based on information calculated during the operation in the fast operation mode. When the operation mode of the comfort operation mode is an overload, the operation mode control unit 190 or 290 may control the air conditioner by controlling the air volume to be a medium air volume or a weak air volume to reduce or maintain the air volume.
On the other hand, when the operation mode of the comfort operation mode is a small load or a standard load, the operation mode control means 190, 290 may control the air conditioner by controlling the air volume to be weak air to reduce the air volume.
In addition, in the operation mode of the comfort operation mode, the operation mode control units 190, 290 may set the wind direction of the wind blowing unit 15 to be upward to blow the wind to a plurality of areas.
As examples of the operation modes before/after the target temperature is reached, there may be divided into a mode (first section) of operating at the air conditioning maximum cooling capacity before the target temperature is reached and an operation mode (second section) of performing the customized control according to the load judged at the reaching time t1 after the target temperature is reached, so that various operation modes are provided by changing the air volume according to the load level estimated at the time t1 to save electricity or cool comfortably.
In which an initial temperature and a target temperature at which the air conditioner starts to operate and variables of various environments occurring during this time are extracted as learning factors. Then, the air conditioner is changed to a comfort mode after the fast mode using the maximum cooling power, and in the process, the operation of the air conditioner can be controlled by selecting an appropriate load so that the user does not feel a temperature increase.
In addition, various environmental factors are used to determine the load, and if an external operation (elevating temperature or adjusting air volume, etc.) controlling the operation of the air conditioner occurs during the operation in the comfort mode, the load can be more accurately estimated by inputting the external operation to the learning units 160, 360 to form a new result node.
The summary is as follows. In the case of a cooling/air conditioning apparatus as an example, the cooling/air conditioning apparatus maintains the temperature at the time of the end of the rapid operation mode in the interval t1-t2 with reference to the end time t1 of the rapid operation mode, and then increases the temperature in the interval t1-t2 at a stepwise rate of change compared with the interval t2-t3 which is temporally continuous with the interval t1-t 2. In the case of a heating air conditioner as an example, the temperature is maintained in the interval t1-t2 with reference to the end time t1 of the rapid operation mode, and then the temperature is decreased in a stepwise manner from the interval t1-t2 in the interval t2-t3 which is temporally continuous with the interval t1-t 2. Here, as shown in fig. 12, in one embodiment, the stepwise change rate means a temperature slow change or a temperature stepwise rise. Of course, the air conditioner may be operated such that the temperature changes on a straight line or a broken line or a curved line with the Temp-a point of t1 as one point, the Temp-aa point of t2 as another point, and then the Temp-ab point of t3 as yet another point.
Then, the time interval of t1-t2, t2-t3 may also be increased or decreased. For example, the air conditioner may receive control signals for raising and lowering the temperature from the interface unit 140, 240 during operation over the interval t1-t 2. The user changes the temperature over the interval t1-t2 as an example. The central control unit 150, 250 may increase the length of time of the t1-t2 interval upon receiving the control signal for decreasing the temperature. For example, one minute may be increased. Conversely, the central control unit 150, 250 may decrease the length of time of the t1-t2 interval upon receiving the control signal for increasing the temperature. For example, one minute may be reduced. The same applies to the case where the control signal is received over the interval t2-t 3. As a result, in the comfort operation mode, the increase and decrease of the section corresponding to the time at which the control signal is generated, such as t1-t2 or t2-t3, in response to the control signal for increasing or decreasing the temperature, can be inversely proportional to the increase or decrease of the temperature.
In addition, according to an embodiment of the present invention, the control units 150 and 250 may store location information of a residential area where presence of indoor people is repeatedly confirmed in a space where the air conditioner is installed. That is, in the entire space to which the wind of the air conditioner is transmitted, information of an area where people are mainly sensed or people are frequently present is stored as location information of a residential area, and the air conditioner can be operated in a concentrated manner with the residential area as an object in a fast operation mode. Therefore, in the fast operation mode, the left and right wind directions of the air blowing unit 15 can be set according to the residential area. In the comfort mode thereafter, the left and right wind directions of the air blowing unit 15 may be set in accordance with the living area in the comfort operation mode based on the position information stored in the control units 150 and 250, for example, the position information including the resident area and intermittently identifying the living area where the indoor person is present. Therefore, energy efficiency can be improved by intensively blowing air to a specific area (residential area) stored in the control units 150 and 250 in the fast operation mode and intensively blowing air to a wider area (living area) including the residential area in the comfort operation mode.
In order for the control units 150, 250 to distinguish between a residential area and a living area, the air conditioner includes an Artificial Neural Network (Artificial Neural Network) that can recognize an area where an indoor person is located in an indoor space divided into a plurality of areas from an image obtained by a camera using an external image, and has learned a position recognition result of the indoor person through machine learning (machine learning), and is used by inputting data in the Artificial Neural Network to divide the plurality of areas into a residential area where air is intensively blown and a living area including the residential area. Such information may be position information (left/right and distance information centering on the air conditioner) provided from an external server.
As shown in fig. 12, the operation mode information calculated by the learning unit with respect to the second section may include time information t2, t3 and temperature information Temp-ab, Temp-aa.
Fig. 13 is a diagram showing the structure of the learning unit according to the embodiment of the present invention. The structure of the learning units 160, 360 of fig. 3 or 4 described above is observed.
The learning units 160, 360 include: an input layer (input) having N parameters as input nodes, an Output layer (Output) having operation mode information as Output nodes, and one or more M hidden layers disposed between the input layer and the Output layer. As an example of the parameter, the factor shown in fig. 5 or fig. 6 may be mentioned, but is not limited thereto.
Here, a weight is set in an edge (edge) connecting nodes of a plurality of layers, and the weight or the presence or absence of the edge may be added or deleted or updated in the learning process. Therefore, the weights of the plurality of nodes and the plurality of edges arranged between the k input nodes and the i output nodes can be updated by the learning process or the interrupt input. As shown in fig. 13, the output node may be configured with i to output 1/0 or the probability equivalent value per pattern. Alternatively, the output node may be configured to output one node requiring an element (+, -or + 10% or-20%) relatively changed from the operation mode of the first interval. Alternatively, t2, t3, Temp-aa, Temp-ab, etc. shown in FIG. 12 may also form the output node.
All nodes and edges may be set to initial values before the learning units 160, 360 perform learning. However, if the information is an accumulated input, the weights of the nodes and edges of fig. 13 change, in the process, a match of the parameters generated in the first section with the operation pattern information suitable for the second section can be formed. Especially in the case of using the cloud server 300, since the learning unit 360 may receive a large number of parameters, the learning unit 360 may perform learning based on a large number of data.
The interrupt input refers to information for indicating this when the wind speed or temperature is changed by the user after the operation mode information on the second section is output. Therefore, if the operation mode information of the second section is calculated after k parameters are input in the first section and then the Interrupt input is received again, the Interrupt input may be input to another node (Interrupt P) by a predetermined value to calculate new operation mode information or update the learning units 160, 360.
In summary, the weights of the nodes and edges between the input nodes and the output nodes forming the learning units 160, 360 of fig. 13 may be updated by the learning process of the learning units 160, 360 or the interrupt input of the central control unit.
Fig. 14 is a diagram showing an exemplary structure of a learning unit according to an embodiment of the present invention. The learning units 160, 360 include five units (units) as inputs, three hidden layers, and three units as outputs. The first hidden layer includes twenty cells, the second hidden layer includes thirteen cells, and the third hidden layer includes five cells. Each node is always configured with a link between them, and the weight of the link can be set.
As the input values, an indoor initial temperature, a target set temperature, an initial N-minute (e.g., three-minute) temperature difference, a rapid interval temperature difference, a time to reach the target temperature may be mentioned. These five input values are combined to map to twenty units, and the weight (weight) and bias (bias) of the link required for these mappings can be continuously learned from the initially set values, and these values can be changed in the learning process.
Likewise, from twenty cells to thirteen cells of the second hidden layer, the thirteen cells of the second hidden layer are mapped again to five cells of the third layer. Then, from the last five units, three output units are connected, each representing a degree of load. That is, each output unit is calculated to be one, which can be connected to a standard load, an overload, a small load, respectively. Of course, depending on the implementation method, the output unit may be one, and the values of the output are set to 0, 1, 2 to calculate the load degree.
The links between each layer or the weights and biases applied to these links may continue to change during the learning process. Alternatively, the information of these learning units 160, 360 may be updated externally to set the weights and biases.
In one embodiment, when comparing the input unit with the unit of the first hidden layer, one hundred weights can be generated for a total of five input nodes, since each input node is set with the weight (weight) values of the nodes of twenty hidden layers. In addition, since the offset values added by addition are calculated after multiplication by the weight values, a total of twenty offset values can be generated. Alternatively, the offset value may be set for each weight.
When determining the weights and biases on the respective links of the learning units 160, 360, if five parameters are input to the learning units 160, 360, the values of 0, 1, or 2 are finally calculated. The calculated value may be provided to the air conditioner as operation mode information for setting a load in a comfort mode of the air conditioner.
In input, the set of weights applied to the links configured at the first hidden layer is { w1_1, w1_ 2.., w1_ i }, and the set of offsets is { b1_1, b1_ 2.., b1_20 }. Here, the value of i may be 20 or more and 100(5 × 20) or less. In fig. 14, i has a value of 100 since links are configured between the input node and all nodes of the first hidden layer. The values of twenty cells of the first hidden layer are input by applying { w1_1, w1_ 2., w1_100} and { b1_1, b1_ 2., b1_20} to the five factors input.
The set of weights applied to the link configured from the first hidden layer to the second hidden layer is { w2_1, w2_ 2.., w2_ j }, and the set of offsets is { b2_1, b2_ 2.., b2_13 }. Here, the value of j may be 20 or more and 260(20 × 13) or less. In fig. 14, since links are configured between all nodes between the first hidden layer and the second hidden layer, j has a value of 260. The values of the twenty cells of the first hidden layer are input as the values of the thirteen cells of the second hidden layer by applying { w2_1, w2_ 2., w2_260} and { b2_1, b2_ 2., b2_13 }.
The set of weights applied to the link configured from the second hidden layer to the third hidden layer is { w3_1, w3_ 2.., w3_ k }, and the set of offsets is { b3_1, b3_ 2.., b3_5 }. Here, the value of k may be at least 13 or more and 65(5 × 13) or less. In fig. 14, k has a value of 65 since links are configured between all nodes between the second hidden layer and the third hidden layer. Values of thirteen cells of the second hidden layer are input as values of five cells of the third hidden layer by applying { w3_1, w3_ 2., w3_65} and { b3_1, b3_ 2., b3_5 }.
The set of weights applied to the link from the third hidden layer configuration to the output is { w4_1, w4_ 2.., w4_ p }, and the set of offsets is { b4_1, b4_ 2.., b4_3 }. Here, the value of p may be at least 5 or more and 15(5 × 3) or less. In fig. 14, since links are configured between all nodes between the third hidden layer and the output, p has a value of 15. The values of the five cells of the third hidden layer are connected into three output cells using { w4_1, w4_ 2., w4_ p } and { b4_1, b4_ 2., b4_3}, thereby calculating the load degree.
Fig. 15 is a diagram showing interaction between the supply air speed and the cooling air and parameters of an air conditioner for discharging the cooling air according to an embodiment of the present invention. In the first section in which the cooling air is discharged, the learning units 160 and 360 may increase the wind speed or the outdoor unit may increase the cooling air in the following cases. The learning unit 160, 360 may output operation mode information for controlling to increase a wind speed or increase the cooling air of the outdoor unit in a case of i) being proportional to an indoor initial temperature TempInit, ii) being inversely proportional to a target set temperature TempTarget, iii) being inversely proportional to a temperature change rate InitRate of an initial section, iv) being inversely proportional to a temperature change rate PowerRate of a first section, or v) being proportional to a time interval PowerTime.
Therefore, when the learning units 160 and 360 are configured of one or more hidden layers, one or more edges and nodes set with high weight or proportional weight may be arranged between the wind speed, the wind volume, or the cooling air and the TempInit/PowerTime. These edges and nodes form a quadrilateral denoted "proportional".
Conversely, more than one edge and node set with a low or inversely proportional weight may be disposed between the wind speed, the wind volume, or the cooling air and the TempTarget/InitRate/PowerRate. These edges and nodes form a quadrilateral denoted as "inverse ratio".
Fig. 16 is a diagram showing the interaction between the supply wind speed and the heated air and the parameters of an air conditioner for discharging the heated air according to another embodiment of the present invention.
In the first section in which the heating air is discharged, the learning units 160 and 360 may increase the wind speed or the outdoor unit may increase the heating air in the following cases. The learning unit 160, 360 may output operation mode information for controlling to increase a wind speed or increase heating air of the outdoor unit in a case of i) being inversely proportional to the indoor initial temperature TempInit, ii) being directly proportional to the target set temperature TempTarget, iii) being inversely proportional to the temperature change rate InitRate of the initial section, iv) being inversely proportional to the temperature change rate PowerRate of the first section, or v) being directly proportional to the time interval PowerTime.
Therefore, when the learning units 160 and 360 are configured of one or more hidden layers, one or more edges and nodes set with high weight or proportional weight may be disposed between the wind speed, the wind volume, or the cooling air and the TempTarget/PowerTime. These edges and nodes form a quadrilateral denoted "proportional".
Conversely, more than one edge and node with low or inverse weighting may be configured between the wind speed, wind volume or cooling air and TempInit/InitRate/PowerRate. These edges and nodes form a quadrilateral denoted as "inverse ratio".
The proportional/inverse ratios and parameters shown in fig. 15 and 16 are merely examples and may be selected in various ways.
Based on fig. 13 to 16, the learning units 160, 360 receive one or more parameters as learning factors, and when the learning factors are different, the output operation mode information may also be different. For example, when a first learning factor input at a first time and a second learning factor input at a second time after the first time are different from each other, the learning units 160, 360 calculate second operation mode information different from the first operation mode information output at the first time after the second time, thereby outputting the operation mode information of the second section differently at the first time and the second time. As a result, the operation mode information of the second section at the first time may be different from the operation mode information of the second section at the second time.
When the embodiments of the present invention are applied, a method of estimating a load based on learning in order to effectively control cooling or heating of an air conditioner and an apparatus applying the same may be provided. In particular, in the embodiment of the present invention, the correlation between the environmental factor before/after the time point at which the target set temperature is reached after the air conditioner is operated and the temperature pattern according to cooling (or heating) is learned, so that the load is estimated by stages after the target temperature is reached, to enable efficient cooling operation. Therefore, even if the same target temperature is set, various environmental factors before the target temperature is reached become elements for controlling the air conditioning operation after the target temperature is reached in different ways.
When the embodiment of the present invention is applied, the air conditioner calculates an operation mode corresponding to the parameter calculated during operation as a learning factor.
When applying the embodiment of the present invention, the cloud server may calculate an operation mode suitable for each air conditioner after learning based on parameters provided by the plurality of air conditioners calculated during operation.
When an embodiment of the present invention is applied, the air conditioner can maintain the target attainment temperature within a predetermined range based on a smaller power consumption amount per hour after operating to achieve the predetermined target attainment temperature.
In the embodiment of the present invention, the power consumption amount per hour of the second section (comfort mode) may be configured to be smaller than the power consumption amount per hour of the first section (fast mode). The first interval is an interval in which electric power is initially used to the maximum extent to perform rapid cooling/heating, and is short in time. On the other hand, the second section is a section in which the level of cooling/heating provided in the first section is maintained, and the time is long. Therefore, the power consumption per hour in the first interval is larger than the power consumption per hour in the second interval. Of course, the time interval of the first interval is also smaller than the time interval of the second interval. For example, the first interval may be set to not more than ten minutes at most, and the second interval may be maintained for 3 hours or 5 hours longer than the first interval.
The second section may be interrupted at a time when the indoor unit is interrupted to provide the cooling or heating function, for example, at a time when the user switches the mode or turns off the power supply to change the operation. Therefore, the initial amount of power consumption in the second interval may be equal to or higher than the amount of power consumption in the first interval. However, when the average power consumption amount per hour K2 is calculated based on the total time of the second section, K2 < K1 is satisfied compared to the average power consumption amount per hour K1 used in the first section.
While the above description describes that all the constituent elements constituting the embodiments of the present invention are combined together or operated in combination, the present invention is not necessarily limited to the above embodiments, and all the constituent elements may be selectively combined to operate in one or more within the scope of the object of the present invention. All the components described above may be implemented as a single piece of hardware, or may be implemented as a computer program having a program module that is configured to execute a part or all of the functions combined in one or a plurality of pieces of hardware, with part or all of the components being optionally combined. Codes and code segments constituting the above computer program can be easily derived by those skilled in the art of the present invention. The Computer program described above may be stored in a Computer Readable storage medium (Computer Readable Media) and read and executed by a Computer, thereby implementing the embodiments of the present invention. The storage medium of the computer program includes a magnetic storage medium, an optical storage medium, a semiconductor storage element, and the like. In addition, the computer program implementing the embodiments of the present invention includes program modules transmitted in real time through an external device.
Although the embodiments of the present invention have been described above, those skilled in the art can make various changes and modifications. Therefore, it is to be understood that the above-described modifications and variations are included in the scope of the present invention without departing from the scope of the present invention.

Claims (16)

1. An air conditioner, comprising:
an air supply unit for discharging air provided by the outdoor unit;
a parameter generation unit that calculates one or more parameters in the first operation mode;
an operation mode control unit that controls the air blowing unit or the outdoor unit based on operation mode information in which a second operation mode is set after a section operated in the first operation mode;
the communication unit is used for receiving and transmitting data with the cloud server; and
a central control unit for controlling the parameter generation unit, the operation mode control unit, and the communication unit,
the first operation mode is operated only within a preset time range, the central control unit instructs the operation mode control unit to operate in the second operation mode after the first operation mode,
the operation mode information is operation mode information received from the cloud server after the communication unit transmits the parameter to the cloud server or operation mode information calculated in a built-in learning unit, the operation mode control unit sets a wind direction and an air volume of the air blowing unit using the operation mode information,
the absolute value of the temperature change rate per hour in the first operating mode is larger than the absolute value of the temperature change rate per hour in the second operating mode,
when the air conditioner is a cooling air conditioner, the temperature is increased at a stepwise rate of change from a first section in a second section temporally continuous with the first section after the temperature at the time of ending the first operation mode is maintained in the first section with reference to the end time of the first operation mode,
when the air conditioner is a heating air conditioner, after the temperature is maintained in a first section with reference to the end time of the first operation mode, the temperature is reduced at a stepwise rate of change from the first section in a second section temporally continuous with the first section,
the first interval and the second interval constitute a time interval of a second operation mode.
2. The air conditioner according to claim 1,
also comprises an interface unit which receives a control signal for elevating the temperature from the outside,
in the case where the air conditioner is a refrigerating air conditioner,
in the second operation mode, the length of the first section or the second section is increased or decreased in inverse proportion to the temperature increase or decrease control signal.
3. The air conditioner according to claim 1,
in the second operation mode, in a case where a difference between a temperature of a space in which the air conditioner is set and a target set temperature is included in a predetermined temperature range set in advance or the difference is maintained for a predetermined time, the central control unit instructs the operation mode control unit to switch to the first operation mode.
4. The air conditioner according to claim 1,
the central control unit instructs the operation mode control unit to switch to a humidity control operation mode when the humidity sensed at the timing of switching from the first operation mode to the second operation mode is equal to or higher than a set reference,
when the humidity reaches a set reference and is maintained for a predetermined time, the central control unit instructs the operation mode control unit to cancel the humidity control operation mode.
5. The air conditioner according to claim 1,
when the central control unit stores the position information of the residential area where the indoor personnel are repeatedly confirmed in the space where the air conditioner is installed, the left and right wind directions of the air supply unit in the first operation mode are set corresponding to the residential area.
6. The air conditioner according to claim 5,
when the central control unit stores the position information including the resident area and intermittently confirming the living area of the indoor person, the left and right wind directions of the air blowing unit in the second operation mode are set according to the living area.
7. The air conditioner according to claim 1,
the central control unit instructs the operation mode control unit to operate the first operation mode if the air conditioner starts to operate and a difference between an indoor temperature and a target set temperature is greater than or equal to a predetermined value or if the air conditioner receives an input signal instructing to drive the first operation mode.
8. The air conditioner according to claim 1,
the parameter includes at least one of an indoor initial temperature at a start time of a section operated in the first operation mode, a target set temperature of the section, a temperature change rate of a preset initial section of the section, a temperature change rate of the section, and a time interval between the start time and the end time of the section.
9. The air conditioner according to claim 1,
the air conditioner further includes a learning unit receiving the parameter as a learning factor to output the operation mode information,
the learning unit includes:
the input layer takes the parameters as input nodes;
an output layer which takes the operation mode information as an output node; and
one or more hidden layers disposed between the input layer and the output layer,
weights of nodes and edges between the input node and the output node are updated by a learning process of the learning unit.
10. A method of driving an air conditioner based on learning, comprising:
a step of operating in a first operation mode within a preset time range and discharging air from an air supply unit of the air conditioner;
a step in which a parameter generation unit calculates one or more parameters in the first operation mode;
a step in which the operation mode control unit controls the air blowing unit or the outdoor unit based on operation mode information for setting a second operation mode after a section operated in the first operation mode; and
a step in which the first operating mode is operated only within a predetermined time range, the central control unit instructs the operating mode control unit to operate in the second operating mode after the first operating mode,
the operation mode information is operation mode information received from a cloud server after the communication unit transmits the parameter to the cloud server or operation mode information calculated in a built-in learning unit,
after the step of indicating, further comprising:
a step in which the operation mode control means sets the wind direction and the air volume of the air blowing means using the operation mode information,
the absolute value of the temperature change rate per hour in the first operating mode is larger than the absolute value of the temperature change rate per hour in the second operating mode,
after the step of indicating,
when the air conditioner is a refrigerating air conditioner, the air conditioner includes:
a step of increasing the temperature at a stepwise rate of change from the first section to a second section temporally consecutive to the first section after the temperature at the time of ending the first operation mode is maintained in the first section with reference to the end time of the first operation mode,
when the air conditioner is a heating air conditioner, the air conditioner includes:
a step of reducing the temperature at a stepwise rate of change from the first section in a second section temporally continuous with the first section after the temperature is held in the first section with reference to the end time of the first operation mode,
the first interval and the second interval constitute a time interval of a second operation mode.
11. The method for driving an air conditioner based on learning according to claim 10,
the air conditioner further includes an interface unit receiving a control signal for elevating a temperature from the outside,
in the case where the air conditioner is a refrigerating air conditioner,
in the second operation mode, the length of the first section or the second section is increased or decreased in inverse proportion to the temperature increase or decrease control signal.
12. The method for driving an air conditioner based on learning according to claim 10, further comprising:
a step in which, in the second operation mode, if a difference between a temperature of a space in which the air conditioner is set and a target set temperature is included in a predetermined temperature range set in advance or the difference is maintained for a predetermined time, the central control unit instructs the operation mode control unit to switch to the first operation mode.
13. The method for driving an air conditioner based on learning according to claim 10, further comprising:
a step in which the central control unit instructs the operation mode control unit to switch to a humidity control operation mode when the humidity sensed at the time of switching from the first operation mode to the second operation mode is equal to or higher than a set reference; and
and a step in which the central control unit instructs the operation mode control unit to cancel the humidity control operation mode when the humidity reaches a set reference and is maintained for a predetermined time.
14. A cloud server, comprising:
a communication unit that receives, from a plurality of air conditioners, one or more parameters calculated in a first operation mode in correspondence with a set temperature set for each air conditioner, and transmits, in correspondence with the one or more parameters, operation mode information for setting a second operation mode to the plurality of air conditioners, respectively;
a learning unit that receives the parameter of a first air conditioner of the plurality of air conditioners as a learning factor and outputs operation mode information for setting a second operation mode of the first air conditioner after a section in which the first air conditioner is operated in the first operation mode; and
a server control unit for controlling the learning unit and the communication unit,
in the operation mode information transmitted to the plurality of air conditioners, an absolute value of a temperature change rate per hour in the first operation mode of each air conditioner is larger than an absolute value of a temperature change rate per hour in the second operation mode of the same air conditioner.
15. The cloud server of claim 14,
the parameter includes at least one of an indoor initial temperature at a start time of a section operated in the first operation mode, a target set temperature of the section, a temperature change rate of a preset initial section of the section, a temperature change rate of the section, and a time interval between the start time and the end time of the section.
16. The cloud server of claim 15,
the learning unit includes:
the input layer takes the parameters as input nodes;
an output layer which takes the operation mode information as an output node; and
one or more hidden layers disposed between the input layer and the output layer,
weights of nodes and edges between the input node and the output node are updated by a learning process of the learning unit.
CN201910284184.1A 2018-04-10 2019-04-10 Air conditioner, cloud server, and method for driving and controlling air conditioner Active CN110360725B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR10-2018-0041550 2018-04-10
KR1020180041550A KR102026020B1 (en) 2018-04-10 2018-04-10 Air-conditioner based on parameter learning using artificial intelligence, cloud server, and method of operating and controlling thereof

Publications (2)

Publication Number Publication Date
CN110360725A CN110360725A (en) 2019-10-22
CN110360725B true CN110360725B (en) 2021-07-23

Family

ID=66102537

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910284184.1A Active CN110360725B (en) 2018-04-10 2019-04-10 Air conditioner, cloud server, and method for driving and controlling air conditioner

Country Status (4)

Country Link
US (1) US11353226B2 (en)
EP (1) EP3553402B1 (en)
KR (1) KR102026020B1 (en)
CN (1) CN110360725B (en)

Families Citing this family (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7215070B2 (en) * 2018-10-23 2023-01-31 富士通株式会社 Control program, control method and control device
EP3885664B1 (en) * 2018-12-12 2023-04-26 Mitsubishi Electric Corporation Air conditioning control device and air conditioning control method
CN111692721B (en) * 2019-03-15 2023-09-22 开利公司 Control method for air conditioning system
CN111692703B (en) * 2019-03-15 2023-04-25 开利公司 Fault detection method for air conditioning system
WO2020218219A1 (en) * 2019-04-26 2020-10-29 ダイキン工業株式会社 Air-conditioning system, machine learning device, and machine learning method
US11761659B2 (en) * 2019-06-24 2023-09-19 Lg Electronics Inc. Method for predicting air-conditioning load on basis of change in temperature of space and air-conditioner for implementing same
CN110736229A (en) * 2019-10-29 2020-01-31 珠海格力电器股份有限公司 Running state control method and device of air conditioner, processor and air conditioning equipment
AU2020392948A1 (en) * 2019-11-26 2022-07-14 Daikin Industries, Ltd. Machine learning device, demand control system, and air-conditioner control system
CN111102676B (en) * 2019-12-16 2021-01-22 珠海格力电器股份有限公司 Air conditioner indoor unit constant air volume control method and air conditioner
KR20210100355A (en) 2020-02-06 2021-08-17 엘지전자 주식회사 Air conditioner and method for controlling for the same
CN111623492B (en) * 2020-05-06 2022-07-12 青岛海尔空调电子有限公司 Air conditioner and compressor control method thereof
CN111831871B (en) * 2020-07-07 2023-10-24 海尔(深圳)研发有限责任公司 Method, device and equipment for recommending air conditioner working mode
CN111765610B (en) * 2020-07-09 2021-07-23 苏州智数家建筑科技有限公司 Indoor humidity control method
CN112032964B (en) * 2020-08-17 2022-03-11 Tcl空调器(中山)有限公司 Air conditioner air deflector closing control method and system and storage medium
CN114526540A (en) * 2020-11-23 2022-05-24 广东美的制冷设备有限公司 Air conditioner, method of controlling the same, and computer-readable storage medium
CN112524767B (en) * 2020-12-07 2022-06-24 佛山市顺德区美的电子科技有限公司 Air conditioner control method, air conditioner control device, air conditioner and storage medium
KR102634138B1 (en) * 2021-01-22 2024-02-05 엘지전자 주식회사 An apparatus and method for controlling an operation of an air conditioner
CN113375305A (en) * 2021-06-01 2021-09-10 青岛海尔空调器有限总公司 Control method and device of air conditioner, electronic equipment and storage medium
CN113280464B (en) * 2021-06-09 2022-03-04 珠海格力电器股份有限公司 Air conditioner control method, air conditioner controller and air conditioning unit
CN113959051B (en) * 2021-10-08 2023-08-15 青岛海尔空调电子有限公司 Control method for air conditioner and air conditioner
CN115164365A (en) * 2022-06-10 2022-10-11 青岛海尔空调器有限总公司 Control method and device of air conditioner and air conditioner
CN115111727A (en) * 2022-06-10 2022-09-27 青岛海尔空调器有限总公司 Control method and device of air conditioner and air conditioner
CN115523640A (en) * 2022-08-29 2022-12-27 青岛海尔空调器有限总公司 Air conditioner control method and air conditioner control system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105757901A (en) * 2016-04-01 2016-07-13 珠海格力电器股份有限公司 Control method and system of air conditioner
CN107355946A (en) * 2017-06-29 2017-11-17 兰州理工大学 A kind of comfortable energy-saving control system of home furnishings intelligent based on Consumer's Experience and method
CN107504656A (en) * 2017-09-15 2017-12-22 珠海格力电器股份有限公司 A kind of air conditioner Learning Control Method and system

Family Cites Families (38)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2522102B2 (en) * 1990-09-19 1996-08-07 ダイキン工業株式会社 Operation control device for air conditioner
JP2611051B2 (en) * 1991-03-29 1997-05-21 シャープ株式会社 Air conditioner
JP2522102Y2 (en) 1992-03-30 1997-01-08 リズム時計工業株式会社 Clock circuit for analog clock
CN1098186A (en) * 1993-03-29 1995-02-01 三洋电机株式会社 The control device of air conditioner
JP3203126B2 (en) * 1994-04-19 2001-08-27 三洋電機株式会社 Control device for air conditioner
US6220517B1 (en) * 1998-04-22 2001-04-24 Denso Corporation Air-conditioning device
KR100488010B1 (en) * 2001-10-11 2005-05-09 엘지전자 주식회사 Control method of Airconditioner
US7784704B2 (en) * 2007-02-09 2010-08-31 Harter Robert J Self-programmable thermostat
KR101558572B1 (en) * 2008-12-23 2015-10-07 엘지전자 주식회사 Method for controlling air conditioner
US8498753B2 (en) * 2009-05-08 2013-07-30 Ecofactor, Inc. System, method and apparatus for just-in-time conditioning using a thermostat
KR101678432B1 (en) * 2009-09-04 2016-11-22 엘지전자 주식회사 Air conditioner and controlling method thereof
WO2011037095A1 (en) * 2009-09-28 2011-03-31 ダイキン工業株式会社 Control device
US8369995B2 (en) * 2010-03-19 2013-02-05 Denso Wave Incorporated Central air-conditioning system
JP5958912B2 (en) * 2011-08-24 2016-08-02 パナソニックＩｐマネジメント株式会社 Heating system control method and heating system
CN103162375B (en) * 2011-12-13 2015-09-30 珠海格力电器股份有限公司 Air-conditioner and control method thereof and device
JP5865784B2 (en) * 2012-06-05 2016-02-17 日立アプライアンス株式会社 Air conditioner
US9696735B2 (en) * 2013-04-26 2017-07-04 Google Inc. Context adaptive cool-to-dry feature for HVAC controller
WO2015108179A1 (en) * 2014-01-17 2015-07-23 株式会社 東芝 Operation parameter value learning apparatus, operation parameter value learning method, and learning-type device control apparatus
KR102262245B1 (en) * 2014-04-30 2021-06-09 삼성전자주식회사 Air conditioner and method for control of air conditioner
TWI607191B (en) * 2014-08-15 2017-12-01 台達電子工業股份有限公司 Ventilation equipment having dirty filter detecting function and detecting method of the ventilation equipment
US10119714B2 (en) * 2014-09-10 2018-11-06 Cielo WiGle Inc. System and method for remotely controlling IR-enabled appliances via networked device
KR101611738B1 (en) * 2014-10-16 2016-04-11 엘지전자 주식회사 An air conditioning system and a method for controlling the same
TWI546506B (en) * 2014-12-04 2016-08-21 台達電子工業股份有限公司 Controlling system for environmental comfort value and controlling method of the controlling system
US9869485B2 (en) * 2015-02-27 2018-01-16 Mitsubishi Electric Corporation System and method for controlling an HVAC unit based on thermostat signals
KR102424689B1 (en) * 2015-05-15 2022-07-26 삼성전자 주식회사 Method and apparatus of heating ventilation air conditioning for controlling start
WO2016186417A1 (en) 2015-05-15 2016-11-24 삼성전자 주식회사 Method for controlling activation of air conditioning device and apparatus therefor
JP6807556B2 (en) * 2015-10-01 2021-01-06 パナソニックＩｐマネジメント株式会社 Air conditioning control method, air conditioning control device and air conditioning control program
US10203127B2 (en) * 2016-04-29 2019-02-12 Trane International Inc. Time-constrained control of an HVAC system
KR101824324B1 (en) * 2016-07-11 2018-03-14 엘지전자 주식회사 Air conditioner and method for controlling thereof
EP3511639B1 (en) * 2016-09-12 2024-01-24 Mitsubishi Electric Corporation System
US11137161B2 (en) * 2017-03-30 2021-10-05 Samsung Electronics Co., Ltd. Data learning server and method for generating and using learning model thereof
KR101958713B1 (en) 2017-08-18 2019-07-02 엘지전자 주식회사 A controlling method of an air conditioner
CN108278732B (en) * 2018-01-04 2019-12-17 珠海格力电器股份有限公司 Air conditioner control method and device, storage medium and air conditioner
KR102069574B1 (en) * 2018-02-02 2020-02-11 엘지전자 주식회사 Air-conditioner based on parameter learning using artificial intelligence, cloud server, and method of operating and controlling thereof
CN108361927A (en) * 2018-02-08 2018-08-03 广东美的暖通设备有限公司 A kind of air-conditioner control method, device and air conditioner based on machine learning
KR102040953B1 (en) * 2018-04-10 2019-11-27 엘지전자 주식회사 Air-conditioner with region selective operation based on artificial intelligence, cloud server, and method of operating thereof
WO2020035910A1 (en) * 2018-08-15 2020-02-20 三菱電機株式会社 Air-conditioning device, control device, air-conditioning method, and program
US11085663B2 (en) * 2019-07-19 2021-08-10 Johnson Controls Tyco IP Holdings LLP Building management system with triggered feedback set-point signal for persistent excitation

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105757901A (en) * 2016-04-01 2016-07-13 珠海格力电器股份有限公司 Control method and system of air conditioner
CN107355946A (en) * 2017-06-29 2017-11-17 兰州理工大学 A kind of comfortable energy-saving control system of home furnishings intelligent based on Consumer's Experience and method
CN107504656A (en) * 2017-09-15 2017-12-22 珠海格力电器股份有限公司 A kind of air conditioner Learning Control Method and system

Also Published As

Publication number Publication date
EP3553402A1 (en) 2019-10-16
US11353226B2 (en) 2022-06-07
EP3553402B1 (en) 2022-12-14
KR102026020B1 (en) 2019-11-26
US20190309970A1 (en) 2019-10-10
CN110360725A (en) 2019-10-22

Similar Documents

Publication Publication Date Title
CN110360725B (en) Air conditioner, cloud server, and method for driving and controlling air conditioner
KR102069574B1 (en) Air-conditioner based on parameter learning using artificial intelligence, cloud server, and method of operating and controlling thereof
CN110360715B (en) Air conditioner selectively operated in area, cloud server and operation method thereof
US7489988B2 (en) Generator control system, generating apparatus control method, program and record medium
JP4719538B2 (en) Air conditioning system
AU2011202347B2 (en) Method for energy saving on electrical systems using habit oriented control
KR101949362B1 (en) Air conditioner and air conditioning system
KR101908311B1 (en) Air conditioner and air conditioning system
CN105042797A (en) Control method for wall-mounted inverter air-conditioner
KR20210093666A (en) Air conditioner and method thereof
CN110726209B (en) Air conditioner control method and device, storage medium and processor
WO2011147297A1 (en) Methods for energy saving on electrical systems using habit oriented control
CN117331320B (en) Household appliance control method and system based on electric energy meter data analysis
KR102183484B1 (en) Air-conditioner of predicting air conditioning load using artificial intelligence, cloud server, and method of operating thereof
KR20190121639A (en) Air conditioning apparatus and air conditioning system having the same
KR101300271B1 (en) Apparatus and method for controlling air conditioner
KR102029250B1 (en) Air-conditioner of predicting air conditioning load based on artificial intelligence, cloud server, and method of operating thereof
KR20210122520A (en) Air conditioner and Method for calculating the quantity of refrigerant in air conditioner
US11761659B2 (en) Method for predicting air-conditioning load on basis of change in temperature of space and air-conditioner for implementing same
CN109708269A (en) Air-conditioning and its electric heater control method, device and equipment
KR101435086B1 (en) Multi-type air conditioner and operating method for the same
KR20020076421A (en) The cooling control method of air conditioner
KR102111409B1 (en) Air conditioner
US20230375202A1 (en) Air conditioner and air conditioner control method
KR20180081886A (en) Air conditioner and Method for controlling it

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant